Sam Altman Signals Readiness to Slow AI Development After Security Incident
OpenAI CEO Sam Altman has indicated a shift in his stance, saying the industry may need to pace AI development to give society time to…
Read more
Desk archive
Category: AI News India
AI product, policy, startup and adoption news with India context.
Anthropic’s Mythos Model Finds Weaknesses in Core Cryptographic Algorithms
Anthropic’s Claude Mythos Preview discovered mathematical weaknesses in HAWK and a reduced version of AES in just days, raising questions about AI’s growing role in…
Read more
Anthropic CEO clarifies stance on open-weight models: no ban, but warns of Chinese AI risks
Dario Amodei says Anthropic never advocated for a ban on open-weight models, but fears China’s authoritarian use of AI. He backs global safety testing and…
Read more
Satya Nadella warns Indian enterprises: Relying on a single AI model could be fatal
Microsoft CEO Satya Nadella says companies that outsource their thinking to one AI lab risk losing control of their data, their prompts, and eventually their…
Read more
Claude shared chats and Artifacts found publicly visible on Google — privacy alert for AI users
A large number of Claude conversations and Artifacts became publicly searchable on Google over the weekend, exposing health records, company documents and children’s personal data.…
Read more
Moonshot AI Releases Kimi K3 Open Weights, Sparking Debate on Distillation and Performance
Chinese AI firm Moonshot AI has released the model weights and infrastructure for Kimi K3, claiming near-frontier performance. However, independent tests reveal significant gaps in…
Read more
ChatGPT Citation Rates Vary Sharply by Topic – Travel Leads, Education Lags
A Similarweb report shows ChatGPT includes external links in 22.6% of travel answers but only 4.8% of education answers. The findings offer a data-driven view…
Read more
OpenAI study finds 43.5% of workers use ChatGPT for tasks outside their own profession
OpenAI analysed over 800,000 work-related ChatGPT messages and found that 43.5% of job-specific queries involve tasks from other professions. The trend, called ‘task crossover’, is…
Read more
Microsoft Launches AI Cybersecurity Model MAI-Cyber-1-Flash, Still Relies on OpenAI for Complex Tasks
Microsoft introduces MAI-Cyber-1-Flash, a compact security model that handles 90% of tasks, cutting costs by 50%, while passing tough cases to GPT-5.4.
Read more
Microsoft enters AI cybersecurity market with dedicated model and agentic platform
Microsoft launched MAI-Cyber-1-Flash, its first cybersecurity AI model, and the Perception agentic platform, aiming to help enterprises defend against AI-driven threats. The tools will be…
Read more